A First Look at Intel's 14nm Fab 42 Manufacturing Facility

[Guest]

Intel is adding a new gem to its network of manufacturing facilities.

A First Look at Intel's 14nm Fab 42 Manufacturing Facility : Read more

 

[rpmrush]

And this is why intel will be a top dog in the mobile sector in a couple of years. 5 Billion+ for the plant! They are not afraid to invest heavily. Intel is a BEAST!

 

[hardcore_gamer]

Intel is taking on these guys:

hot electron effect
impact ionization
velocity saturation
drain induced barrier lowering
surface scattering
punchthrough
sub-threshold conduction

 

[eklerus]

omg Intel look at the cpu die size in the 3rd picture it's to big for me ^^

 

[egidem]

And this is precisely why anyone [competition] that continues to doubt or underestimate Intel's prowess is a fool. Something to take from all this is that Intel is not afraid to heavily invest or spend large amounts of money on R&D. If they need to invade the mobile market sector and need to spend $5B on building their facility to make those 14nm processors, then so be it!

I'm curious to see what this situation between Intel and ARM will be like in say 5 years from now. Something tells me that history will repeat itself but this time Intel will be owning ARM.

 

[GreaseMonkey_62]

I didn't know Intel had plants in the US. Starting to make me consider them more. Come on AMD, pull out a big win this year.

 

[dontknownotsure]

[citation][nom]eklerus[/nom]omg Intel look at the cpu die size in the 3rd picture it's to big for me ^^[/citation]

its next step, 30m process

 

[nikorr]

Can't wait.

 

[ojas]

I wonder, does this mean that haswell will have a mid 2013 launch?

Sandy B -> Jan 2011
Ivy B -> April 2012
Haswell -> July 2013 ???

 

[neoverdugo]

The only 2 things intel needs to do are:
1) Drop its price tag
2) Retire the x86 and create a new architecture

Anything else?

 

[digiex]

2) Retire the x86 and create a new architecture

Did you ever met Itanium?

 

[serendipiti]

[citation][nom]neoverdugo[/nom]The only 2 things intel needs to do are:1) Drop its price tag2) Retire the x86 and create a new architectureAnything else?[/citation]

Sadly, final prices for intel products depend on AMD and ARM...

Dropping X86 which could be the real differentiating feature against ARM devices it's a little difficult too.
But probably MS, Linux and open source and cloud / web applications may help.

Anyway, I think it is too soon for that change. The instruction extensions model it is still valid, and it will make more sense doing that rebuild from scratch for the whole architecture the next nodes... (perhaps when 14nm will be mainstream...) Then it will make sense, because you probably will get SoC with GPUs, DisplayPort Bandwidth (PCIe or whatever) ports (to feed SSDs, 10Gb ethernet,...) and keep in mind SATA3 will be deprecated, and useless: USB for external "slow" storage and Display Port "like" (Universal Port, then?) devices for speed. Anything in between will render useless.

The evolution of GPUs (and its integration in a SoC) will also help to start making sense a new architecture.

At the same time, this years of multicore processing will start to show off on the software side.

Perhaps then it will make sense (and will pay its benefits...)

 

[eddieroolz]

Does Intel build a new fab for every process node? It certainly seems that way from my impressions.

 

[rpgplayer]

[citation][nom]eklerus[/nom]omg Intel look at the cpu die size in the 3rd picture it's to big for me ^^[/citation]
[citation][nom]dontknownotsure[/nom]its next step, 30m process[/citation]
It'll still put out half the heat of one Fermi chip....

 

[maxinexus]

When we turn to dust, Intel will still be here standing tall.

 

[mortsmi7]

I could see an architecture change once the size threshold is met. If you can't make it smaller, it's a good time to make it different.

 

[Travis Beane]

High prices for Intel CPUs is part of the reason they can afford to build facilities such as these. If they sold the chips at cost, we'd save a lot of money but progress would stop.

 

[willard]

Why are people calling for an end to x86? What exactly is it that you think is so bad about the instruction set? Current development tools and operating system kernels are very mature. Changing the instruction set would require abandoning all that, and for what?

Methinks the people calling out x86 don't really understand what it really is, just a set of opcodes. Different is not better by default.

 

[alidan]

here let me do some rough math based on a 34nm Intel 510 Series (Elm Crest) SSDSC2MH120A2K5 2.5" 120GB SATA III MLC Internal Solid State Drive (SSD) at 279$

now we can assume that a waffer costs at base 50k, so under current numbers they can yeild 21505gb a waffer
now to make the math easier, i added 1nm to each process, but a shrink from 34nm to 14nm means that they can make 5.44gb of 14nm for ever 1gb of 34nm

crunching the same figures comes out to 116987gb per waffer, and it comes in at .42 cents a gb
but lets not stop there
a 300mm waffer has 282600 mm of working area by my math (max possible, not whats actually used)
a 450mm waffer has 635850 mm of working area by my math (max possible again)

that is a working size of 2.25 the size of the old.

this means that each waffer can hold 263221gb of data and comes to .18cents a gb

that's assuming that the 450nm also costs 50k a waffer, my bets is it costs a bit more, but they wouldnt make it bigger if it cost to much extra so a 14nm ssd will cost between 44 cents to 18 cents a gb. awesome.

useing my same math for this, and assumeing the two best case scenerios on cost, the current top of the line cpu intel makes (Intel Core i7-3960X Extreme Edition Sandy Bridge-E 3.3GHz (3.9GHz Turbo) LGA 2011 130W Six-Core Desktop Processor BX80619i73960X) would cost 85.78$

 

[freggo]

[citation][nom]eklerus[/nom]omg Intel look at the cpu die size in the 3rd picture it's to big for me ^^[/citation]

Hey eklerus, thanks for making me start the day with a good laugh !
Did not 'see' that until I read your one-liner.

 

[acadia11]

All your base are belong to us!!!! Sorry, AMD!!!!

 

[acadia11]

[citation][nom]egidem[/nom]And this is precisely why anyone [competition] that continues to doubt or underestimate Intel's prowess is a fool. Something to take from all this is that Intel is not afraid to heavily invest or spend large amounts of money on R&D. If they need to invade the mobile market sector and need to spend $5B on building their facility to make those 14nm processors, then so be it!I'm curious to see what this situation between Intel and ARM will be like in say 5 years from now. Something tells me that history will repeat itself but this time Intel will be owning ARM.[/citation]

ARM doesn't make chips, they produce designs. It's an IP company. I think ARM will keep on keeping on.

 

[DRosencraft]

It is good that a company is willing to put in the time and cash for R&D, and better for the US that Intel is tryng to add more work in the states. But, I'll wait until they actually produce results from this. Because money is not the only factor that goes into good design and producing a good product. They could just as easily be blowing a ton of money on a project that will ultimately fail. Every company risks this of course when they go into any project, I'm just saying that we shoudn't look at the big dollar signs on a big building and think that's an automatic win. They still have to make this thing pay off in terms of producing something of value.

 

[1foxracing]

So has anyone actually received a CPU marked "Made in USA" I've built several Intel based computers and I've never run across one. I do notice that Mushkin SSD's are tagged " Assembled in USA"

 

[ngoy]

[citation][nom]GreaseMonkey_62[/nom]I didn't know Intel had plants in the US. Starting to make me consider them more. Come on AMD, pull out a big win this year.[/citation]

Really? Most of Intel's main fabs are in the US. Arizona, California, Massachusetts, New Mexico, Oregon. I've been to all the US sites except for the Massachusetts one. There are a few in other countries, I remember Ireland and Israel had facilities, but dont recall the size or process.

Sorry to burst your bubble, but AMD is not pulling anything magic out of any hat soon without moving to a smaller process node in addition to fixing their underperforming bulldozer architecture, and stop trying to blame the OS. Maybe there is a major change coming in their cpus for this year, but with Intel coming out with Ivy Bridge this year on 22nm and 14nm no that far off, they are going to be continously playing catch-up vs leading the pack.